1. Field of the Invention
The invention relates to optical disk drives, and more particularly to processing of wobble signals thereof.
2. Description of the Related Art
The data of Digital Versatile Disks (DVD) and Compact Disks (CD) are encoded and recorded on a single spiral track covering the surface of the disks. If the optical medium is recordable, this spiral track contains a slight sinusoidal deviation from a perfect spiral, wherein the sinusoidal deviation is used to encode modulated address information and referred to as a “wobble signal”. The frequency of the sine curve of the wobble signal is the wobble carrier frequency, and each format of optical disk has the same or different carrier frequency. For example, DVD-R or DVD PAM have wobble carrier frequency of 140.6 kHz, and DVD+R has a wobble frequency of 817.4 kHz.
To extract data recorded on the optical disk, the optical disk drive first detect the wobble signal on the optical disk with a wobble detection circuit. The design of the wobble detection circuit can thus greatly affect performance of an optical disk drive. An optical disk drive reads the wobble signal by detecting the reflection strength of a laser beam moved along the spiral track. FIGS. 1a˜1d are the signals detected by a pick-up head of the optical disk drive. FIG. 1a is a wobble signal without data information recorded thereon, and the waveform of the wobble signal is like a sinusoidal wave. After data information is recorded on disk, the wobble signal is no longer a sinusoidal wave. An ordinary pick-up head scans a track with four photodetectors A, B, C, and D simultaneously. FIGS. 1b and 1c respectively show the exemplary synthesized signals SAD and SBC with recorded data waveform, wherein the signal of FIG. 1b is detected by photodetectors A and D and designated as signal SAD, and the signal of FIG. 1c is detected by photodetectors B and C and designated as signal SBC. Because the wobble carrier components of the signals SAD and SBC are phase-inverted, the recorded data information is obtained by adding signals SAD and SBC. The wobble carrier component is obtained by subtracting amplified signal of SBC from amplified signal of SAD, as shown in FIG. 1d. The subtraction signal is a wobble signal with uncanceled high frequency noise.
FIG. 2 is a block diagram of a conventional wobble detection circuit 200 detecting an absolute time in pre-groove (ATIP) information. ATIP is a method for modulating address information of wobble signals of optical disks such as CD-R or CD-RW. Because only a portion of a wobble signal W0 within a frequency range carries significant information, a band pass filter 202 first filters the wobble signal W0 to obtain a filtered wobble signal W1. An analog to digital converter 204 then converts the analog filtered wobble signal W1 to a digital wobble signal D. An ATIP detector 206 then extracts ATIP information from the digital wobble signal D, and a phase locked loop 208 locks the phase of the digital wobble signal D to obtain a clock signal [not shown] with the same frequency as the digital wobble signal D.
FIG. 3 is a block diagram of a conventional wobble detection circuit 300 detecting an address in pre-groove (ADIP) information. ADIP is generated by modulating address information of wobble signals of optical disks such as DVD+R or DVD+RW. Because only a portion of a wobble signal W0 within a frequency range carries significant information, a band pass filter 302 and a low pass filter 312 respectively filter the wobble signal W0 to obtain filtered wobble signals W2 and W1. Analog to digital converters 304 and 314 then convert the analog filtered wobble signal W2 and W1 to digital wobble signals D2 and D1. An ADIP detector 306 then extracts ADIP information from the digital wobble signal D1, and a phase locked loop 308 locks the phase of the wobble signal D2 to obtain a clock signal [not shown] with the same frequency as the digital wobble signal D2.
The band pass filter 202 of FIG. 2 and the band pass filter 302 of FIG. 3 are analog band pass filters. Analog band pass filters have complex circuit structure and require significant chip area to accommodate the circuits thereof. The chip area occupied by a conventional analog band pass filter usually exceeds half of the total chip area of a wobble detection circuit. Additionally, analog band pass filters require high current for filtering the analog wobble signals, consuming considerable energy. Thus, a wobble detection circuit with a digital band pass filter is desirable.
FIG. 4 is a block diagram of a conventional circuit 400 for detecting the wobble carrier frequency. The wobble signal shown in FIG. 1d is first delivered to an automatic gain control module 402, which amplifies the wobble signal to an appropriate strength level. A band pass filter 404 processes the amplified wobble signal by filtering out undesirable out-band noise. The wobble signal is converted to a binary data stream by a binary converter 408 after the direct current (DC) offset of the processed wobble signal is canceled by a high-pass filter 406. Thus, the edge counting module 410 can first detect the edge then simply count the number of edge in a predetermined period to obtain the wobble carrier frequency.
The wobble carrier frequency obtained by the edge counting module 410, however, may be erroneous due to noise in the wobble signal shown in FIG. 1d. Although the band pass filter 404 filters out the noise of the wobble signal, not all of the noise is filtered out. The noise left in the wobble signal may interfere with conversion by the binary converter 408 thus generating an erroneous binary data stream. Thus, the wobble carrier frequency obtained by counting the edges of the erroneous binary data stream is also an erroneous wobble carrier frequency, which is difficult to detect accurate wobble frequency. In addition, the band pass filter 404 is an analog band pass filter, which is complicated and occupying a large chip area.
Significant information, such as address information, is pre-grooved on optical disks in the form of wobble signals. To extract address information from wobble signals, the wobble signals should be appropriately amplified to a predetermined signal level. Automatic gain controllers (AGC) are therefore used in wobble detection circuits to control gain for amplification of the input signals.
Conventional AGC of wobble detection circuits are analog circuits. The analog AGC, however, require capacitors to provide sufficient circuit capacitance to lower the bandwidth of the automatic gain controllers. Because on-chip capacitors with requisite high capacitance occupy considerable chip area, such capacitors are often located externally. Coupling between the analog automatic gain controllers and the external capacitors, however, requires extra IO pins, increasing the cost of PCB.
Some AGC of wobble detection circuits are implemented with digital circuits to avoid capacitance problems. FIG. 5 is a block diagram of a digital automatic gain controller 500. The digital automatic gain controller 500 includes an analog variable gain amplifier 510, an analog to digital converter (ADC) 504, an envelope detection module 502, a digital control module 506, and a digital to analog converter 508. The analog variable gain amplifier 510 amplifies an input signal digital SI according to a gain signal M′ to obtain an amplified signal SI′, which is converted to a digital signal So by ADC 504. The envelope detection module 502 then detects the envelope E of the digital signal So. The digital control module 506 then determines a gain signal M according to the envelope E, and the digital to analog converter converts the digital gain signal M to the analog gain signal M′ to control amplification of the analog variable gain amplifier 510. Thus, the signal gain M of the digital automatic gain amplifier 500 is digitally determined by the digital control module 506 and does not require the high capacitance of analog automatic gain controllers.
Because the input signal SI contains high frequency noise which is induced by recorded data or write pulse, the frequency of the amplified signal SI′ is as high as the frequency of the input signal SI. To meet the requirement of Nyquist sampling theorem, the analog to digital converter 504 convert the amplified signal SI′ to the digital signal SO with a sampling frequency higher than twice times of the highest frequency of recorded data frequency band. Additionally, signal resolution of the envelope E should be high enough that the digital control module 506 can finely adjust the gain signal M according to the envelope E. Thus, the analog to digital converter 504 generate the digital signal SO with high signal resolution. The high sampling rate and high signal resolution of the signals SO, E, and M complicates signal processing and circuit structure of the analog to digital converter 504, the envelope detection module 502, the digital control module 506, and the digital to analog converter 508, greatly increasing the hardware cost of the digital automatic gain controller 500. Thus, a digital automatic gain controller with simpler signal processing is desirable.
There are different ways for addressing optical disks when data is written thereto. If an optical disk is DVD+R or DVD+RW, Address In Pre-groove (ADIP) information records the address of every track zone of the optical disks, allowing the optical disk drive to locate track zones during data recording. If an optical disk is DVD-R or DVD-RW, pre-pit information recorded on the land area of the optical disks is used for addressing track zones of the optical disks during data recording. Thus, a method for demodulating the ADIP information or decoding the pre-pit information is required for optical disk drives to record data onto the optical disks.
The ADIP information is modulated and recorded on the optical disk in the form of a wobble signal. According to DVD+R and DVD+RW specification, each data block of an optical disk includes 93 wobble cycles including eight wobble cycles for storing the ADIP information. Each of the eight wobble cycles can be a negative wobble with an inverted phase or a positive wobble with a normal phase, and different permutations of the eight wobble cycles represent different ADIP symbols. There are only three types of ADIP symbols, synch, data 0, and data 1, respectively represented by one permutation pattern of the negative and positive wobble cycles, wherein synch is an abbreviation for synchronous information.
FIG. 6A shows a wobble signal 610 carrying an ADIP synch symbol. The wobble signal 610 includes eight wobble cycles, comprising four negative wobble (NW) cycles followed by four positive wobble (PW) cycles. If a negative wobble cycle is converted to an ADIP bit 1 and a positive wobble cycle is converted to an ADIP bit 0, the wobble signal 610 is converted to a series of ADIP bits with the permutation of “11110000”. FIGS. 6B and 6C show wobble signals 620 and 630, respectively carrying an ADIP data 0 symbol and an ADIP data 1 symbol. The wobble signal 620 includes eight wobble cycles, one negative wobble cycle followed by five positive wobble cycles further followed by two negative wobble cycles, and can be converted to a series of ADIP bits with the permutation of “10000011”. Similarly, the wobble signal 630 includes eight wobble cycles, one negative wobble cycle followed by three positive wobble cycles followed by two negative wobble cycles further followed by two positive wobble cycles, and can be converted to a series of ADIP bits with the permutation of “10001100”.
FIG. 7 shows a conventional process of demodulating a wobble signal carrying ADIP information into ADIP symbols. The wobble signal to be demodulated is shown in the second row of FIG. 7. A reference wobble with the same frequency and phase as a fundamental frequency and phase of the positive wobble cycle of the wobble signal is shown in the first row of FIG. 7, and the phase difference between the wobble signal and the reference wobble is measured as the phase difference shown in the third row of FIG. 7. Because the reference wobble is generated according to the positive wobble cycle of the wobble signal, the phase difference signal indicates the place where the negative wobble cycle appears, and a series of ADIP bits shown in the fourth row of FIG. 7 can thus be generated according to the phase difference signal with a slicer. The series of ADIP bits is then respectively compared with the three permutation patterns corresponding to the ADIP synch symbol, ADIP data 0 symbol, and ADIP data 1 symbol. Because the series of ADIP bits are “10000011” which agrees with the permutation pattern of ADIP data 0 symbol, a signal indicating the appearance of the ADIP data 0 symbol is enabled, as shown in FIG. 7.
Although the conventional method shown in FIG. 7 is simple, the wobble signal may sometimes carry noise, affecting the generation of phase difference signal. If an erroneous phase difference signal is obtained, the slicer derives erroneous ADIP bits from the erroneous phase difference signal. Because no permutation pattern can agree with the erroneous phase difference signal, no ADIP symbols are generated as the final demodulating output, causing demodulation error. Thus, a method with higher noise tolerance is required to demodulate ADIP symbols.
Optical disks with the format of DVD-R or DVD-RW are addressed according to pre-pit information. According to the DVD-R/RW specification, each error correction code (ECC) block includes 16 sectors, each of which further includes 26 frames. The 26 frames are divided into even frames and odd frames, and each frame includes eight wobble cycles. Every two frames have three pre-pit bits for storing addressing information. FIG. 8 shows the pre-pit bits associated with a wobble signal 800 of two successive frames 802 and 812, wherein the frame 802 is an even frame and the frame 812 is an odd frame. The three pre-pit bits associated with the two frames 802 and 812 may appear on the first three wobble cycles 804, 806, and 808 of the even frame 802 or the first three wobble cycles 814, 816, and 818 of the odd frame 812.
The three pre-pit bits associated every two frames may represent even synch, odd synch, data 0, and data 1 symbols. FIG. 9 shows the information contents of four types of pre-pit symbols recorded with three pre-pit bits. If the pre-pit symbol represents synchronous information appearing on an even frame, the three pre-pit bits are “111”. If the pre-pit symbol represents synchronous information appearing on an odd frame, the three pre-pit bits are “110”. If the pre-pit symbol represents data 1, the three pre-pit bits are “101”. If the pre-pit symbol represents data 0, the three pre-pit bits are “100”. When a pre-pit bit is “1”, the wobble cycle associated with the pre-bit bit has a spike pulse on the peak of the wobble cycle. Otherwise, when an pre-pit bit is “0”, the wobble cycle associated with the pre-bit bit has no spike pulse on the peak of the wobble cycle. Thus, the pre-pit bits can be determined by detecting the spike pulse on the corresponding wobble cycles of two successive frames, and pre-bit symbols can then be determined according to pre-pit bits.
The conventional method for determining pre-pit bits, however, can cause significant errors if noise occurs in the wobble signal carrying pre-pit bits. This can generate erroneous pre-pit bits, and further determine erroneous pre-pit symbols. Thus, a method with higher noise tolerance is required for decoding pre-pit symbols.
In addition, conventional blank area detection of an optical disk drive is realized by detecting the transient spacing of a binaries RF signal. The RF signal is first generated by the optical pickup head. Before binarizing the RF signal to the binaries RF signal, a high pass filter is utilized to remove low frequency components in the RF signal to generate a filtered RF signal. Then the filtered RF signal is binarized to form the binaries RF signal via a slicer with a reference threshold value. Since the amplitude levels of the RF signal from various disks are diverse, it is hard to determine each threshold value to slice RF signal for different disk types. Thus, a method and an apparatus to detect the blank area disregarding the amplitude levels of the RF signal are desired for optical disk drives.